Dynamic control to stabilize the passive robot subjected external perturbance

Researchers are becoming increasingly intrigued by passive walking because of its highly human-like gait and low energy consumption in biped robots. However, most research results based on the assumed ideal walking environment offer only a limited reference for the actual development and application. To tackle this issue, a dynamic model of the passive robot walking in a disturbed environment is constructed in this paper. Additionally, PD control and iterative learning control are employed to regulate the walking of the passive robot under the influence of disturbance, and several analyses could be derived from the simulation results. Firstly, the control method’s effectiveness can be manifested by the ability of the passive robot with disturbance to maintain a stable gait. Secondly, different types of disturbances and their influence on control effectiveness are discussed. The simulation results demonstrate that the control effect is not affected by disturbance types, while large disturbance magnitudes will weaken the walking stability. Furthermore, the gain coefficients of the controller and the influences of different controllers on the control performance are also analyzed and it is concluded that a larger gain parameter of the controller will lead to a better control effect. These conclusions provide a more realistic perspective on the stability control of passive gait and pave the way to the directions for further research.